Guangzhou Nutrition and Health Study (GNHS)

Purpose: The Guangzhou Nutrition and Health Study (GNHS) aims to assess the determinants of risk of metabolic diseases and changes in their relevant indices (e.g., osteoporosis, atherosclerosis, type 2 diabetes, hypertension, metabolic syndrome, non-alcoholic fatty liver disease, cardiovascular diseases, chronic kidney disease, body composition, lung function, cognition function, etc.) in nutritional aspects, as well as other environmental and genetic factors.

Study design: GNHS is a community-based prospective cohort study. Participants: In this cohort, the original GNHS and another cohort study (the controls of a case-control study of hip fractures, CCFH) have been integrated into the one GNHS project. About 4048 apparently healthy residents (aged 40-80 years) in the original GNHS were recruited between 2008 and 2013, and 1070 participants in the CCFH baseline (52-83 years old) were recruited from 2009 to 2015, all living in Guangzhou city (South China) for >5 years. After completing the baseline examination, a total of 5118 participants were recruited during 2008-2015 in the GNHS project.

Visits and Data Collection: Participants were/will be visited every three years by invited to the School of Public Health, Sun Yat-sen University. At each visit, face-to-face interviews, specimen collection, anthropometric measurements, DXA scanning, ultrasonography evaluation, vascular endothelial function evaluation, cardiopulmonary exercise testing, MRI, 14-d real-time continuous glucose monitoring tests, laboratory tests, and multi-omics data were/will be conducted. Up to December 2022, 3442 and 2895 subjects completed the 2nd and 3rd visits. The 4th visit began in 2017 and has been ongoing, and 2243 participants have been revisited so far. About 1500 participants responded in 2020-now at the 5th visit. It is planned to follow up the participant in person for at least 15 years.

Key variables:

• Questionnaire interviews: Structured questionnaires were/will be used to collect the participants' socio-demographic characteristics (e.g., age, sex and household income), lifestyle factors (smoking, passive smoking, alcohol drinking, tea drinking, physical activity), menstruation and reproductive history (women only), sleep quality (Pittsburgh Sleep Quality Index, PSQI), family history, psychological health (Self-Rating Anxiety Scale, SAS), Simplified Geriatric Depression Scale (GDS), social support and participation, cognitive function (Mini-Mental State Examinations (MMSE), and Addenbrooke's cognitive examination (ACE)), habitual dietary intake (a validated 79-item quantitative food frequency questionnaire), use of supplements and history of chronic diseases.
• Physical examinations: Anthropometric measurements (weight, height, waist, hip and neck circumference, etc.), blood pressure tests, handgrip strength, muscle function and bracelet motion monitoring.
• DXA scanning: A dual-energy x-ray absorptiometry (DXA, Discovery W; Hologic Inc.) was/will be used to determine bone density and bone mineral content at the whole body, lumbar spine (L1-L4), left hip sites, bone geometry information at the hip, fat mass and muscle mass at total body and its sub-regions.
• Ultrasonography evaluations: Ultrasonography evaluation of the carotid artery and upper abdominal organs (e.g., liver and kidney) was/will be performed to determine carotid artery intima-media thickness and plaque, fatty liver.
• Vascular endothelial function evaluation: Blood flow-mediated vasodilation.
• Cardiopulmonary exercise testing: Lung function.
• MRI: Magnetic resonance imaging in the brain was/will be used to study brain tissue's microstructure and investigate brain function. Upper-abdomen MRI was/will be conducted to assess the structure and contents of fat and iron of the liver, fat and muscle mass, and vertebral bone marrow fat, and help to identify renal angiomyolipoma and malignant renal tumours.
• 14-d Real-time continuous glucose monitoring tests: A 14-d continuous glucose monitoring was used to determine glycemic responses to various usual daily foods (by a 7-d image-based food diary) using 3-type standard breakfast as internal calibrators.
• Specimen collections: Overnight fasting blood sample was/will be collected and separated into serum, plasma, erythrocyte and leukocyte within two hours. Early morning first-void urine, faces, and saliva samples were/will be collected, separated and stored at -80°C till tests.
• Laboratory tests:

• Fasting serum lipid profile.
• Diabetes-related indices; fasting glucose, insulin, HbA1c, and fructosamine.
• Uric acid.
• Nutritional indices: fatty acids, vitamins, minerals, alkaloids, carotenoids, flavonoids, sulfur-containing amino acids and so on.
• Inflammatory cytokines.
• Index of oxidative stress.
• Adipocytes.
• Sexual hormones.
• Liver and renal function-related markers.
• Routine blood test.
• Multi-omics data:

• Genotyping data (Illumina Asian Screening-750000 arrays).
• Gut microbiota: 16S ribosomal RNA, metagenome, and internal transcribed spacer 2 (ITS2) sequencing.
• Untargeted serum and fecal proteomics: about 430 unique human protein groups in serum, 1253 human protein and 83683 microbial proteins in feces.
• Targeted serum and fecal metabolomics: amino acids, benzenoid, bile acids, carbohydrates, carnitines, fatty acids, indoles, nucleosides, organic acids, organooxygen compounds, phenylpropanoic acids, pyridines and so on. Approximately 200 metabolites have been quantified so far.
• Morbidity and mortality: Relevant data were/will be also retrieved via local multiple health information systems.
• Others: Many other laboratory tests or instrument tests will be developed depending on needs and resources in future.

Data analysis: Analyses of variance and covariance, or mixed effects model were/will be used to compare the mean differences in continuous outcomes (e.g., changes of bone mineral density, body fat mass, or intima-media thickness) among the quartiles. Cox proportional hazards or logistic regression models were/will be used to assess the risk of exposures (e.g., nutrition intakes and physical activity) on categorical outcomes. Path analysis was/will be used to assess the potential mediating effects in the causal pathway between exposures and outcomes.

Inclusion criteria

• Age: 40-80 years (the original GNHS) and 40-85 years (CCFH) at baseline;
• Living in Guangzhou for at least five years;
• Chinese.

Exclusion criteria

• Had a history of hospital-confirmed diabetes, failure(s) of heart, liver, or kidney, cancer, cardiovascular events, metabolic bone diseases, glucocorticoid use (over 3 mo.) or sexual hormone use (over 6 mo.), spine or hip fractures;

On special diet due to a disease or weight control;

• Mental and physical disability;
• Likely to move to other city within 5 years;
• Did not want to attend any one item of the survey or sample collection.